Cervical cancer: Radiotherapy and hyperthermia

Abstract

Background: For many years, the standard treatment of advanced cervical cancer has been radiotherapy (RT), including brachytherapy. The achievement of locoregional tumour control is essential for cure. Results of RT in early stages are reasonably satisfactory, but locoregional failure rates for stage IIIb and IVa are high. In several randomized trials, the addition of hyperthermia (HT) to RT has been investigated.

Randomized trials: The Dutch Deep Hyperthermia Trial was completed in 1996. In this trial a beneficial effect of additional hyperthermia was clearly demonstrated. Three-year locoregional control and overall survival rates were significantly higher in the RT + HT group than in the RT alone group, while radiation toxicity was not affected. Cost-per-life-year-gained was less than 4000 Euros. The results of this trial have led to the acceptance of RT plus HT as standard treatment for advanced cervical cancer in the Netherlands.

 Five trials conducted in Asia have been published, of which three showed significant better complete response, locoregional tumour control and/or disease-free survival rates. One trial showed a trend of better locoregional tumour control and one did not show any benefit.

Conclusion: Hyperthermia added to standard radiotherapy of locally advanced cervical tumours results in considerable therapeutic gain and is cost-effective. For a beneficial effect, the use of an adequate heating technique is an important requirement.

• Radiotherapy
• hyperthermia
• cervical cancer
• randomized study
• pelvic control
• overall survival
• cost-effectiveness

Introduction

The results of radiotherapy in early stages of cervical carcinoma are reasonably satisfactory, but locoregional failure rates of stage II–IV range between 41–72% [1], [2]. Pelvic tumour control is a pre-requisite for cure. A pelvic failure generally indicates a fatal course of the disease and will cause major problems. The potential gain in survival by definitive eradication of locoregional tumour in patients with cervical cancer was estimated to be 50–60% [3], [4].

The median percentage of pO₂ measurements with a level <5 mm Hg (the hypoxic fraction) in cervical tumours was found to range between 42–61%, which may explain the ineffectivity of radiotherapy in larger tumours [5–7]. Previous attempts to improve local control rate by tackling the hypoxic cell population include the application of radiotherapy (RT) under hyperbaric oxygen or in combination with radiosensitisers, both without demonstrating convincing improvements [8], [9]. Another approach is to combine RT with hyperthermia (HT). Besides many phase II studies, six randomized trials comparing the results of RT alone with RT plus HT have been published.

Randomized trials on addition of hyperthermia to radiotherapy

The Dutch Deep Hyperthermia Trial (DDHT)

In this trial, two separate studies were combined [10], [11]. Eleven RT institutes, among which three had HT facilities, participated. Patients with advanced rectal, bladder or cervical cancer were included. The majority of the 114 patients with cervical cancer had tumours with unfavourable prognostic factors: 80% had a FIGO stage IIIb or IVa tumour, 74% had positive pelvic lymph nodes and in 77% the tumour diameter was 6 cm or larger. External-beam RT was given in fractions of 1.8–2 Gy to a total dose of 46–50.4 Gy, prescribed to the tumour and regional pelvic nodes, with an additional boost to residual parametric tumour. A brachytherapy boost was, if feasible, delivered to a total dose of 17 Gy high-dose-rate (38 patients) or 20–30 Gy low-dose-rate (53 patients) in point A. The median overall treatment time was 49 days. Hyperthermia was applied with radiative electromagnetic techniques, once weekly to a total of five treatments. Treatment duration was 60 min after intra-tumour temperature had reached 42°C, with a maximum of 90 min. Power output was increased up to the patient's tolerance, while painful hotspots were avoided by adjustment of treatment settings. The average intra-vaginal temperature was 40.1°C. The DDHT showed significant better complete response, local control and overall survival rates in the RT + HT arm compared to the RT-alone arm, but the addition of HT seemed to be most important for the patients with cervical cancer. In these patients, complete response was 83% in the RT + HT arm and 57% in the RT arm (p = 0.003). Three-years locoregional control rates were 61% and 41%, and overall survival rates 51% and 27% in the RT + HT group and the RT group, respectively (p = 0.009). No differences in acute or late toxicity were found, except for thermal burns in 14% of the patients treated with HT. Besides clinical outcome, cost-per-life-year-gained was evaluated. The evaluation included the operating cost of a HT unit, extra travelling expenses for patients and cost of care in the period after primary treatment. Conservatively assuming that the better local control would lead to an absolute difference in overall survival of 11%, the maximum discounted cost-per-life-year-gained is Euro 3956.

Asian trials

Five Asian randomized trials investigating the addition of HT to RT in patients with advanced cervical cancer have been published [12–16]. The majority of the patients in these studies had stage IIIb tumours. Three studies report significantly better results following RT plus HT compared to RT alone (Table I). In one study, the 2-year pelvic control tended to be higher in the combined treated group [13]. In the most recent study, no beneficial effect from HT was found [16].

Table I.  Results of randomized studies comparing RT with RT + HT.

		Complete response		Pelvic control		Disese-free survival		Overall survival
References	Figo stage and numbers of patients	RT	RT + HT	RT	RT + HT	RT	RT + HT	RT	RT + HT
					2 years		2 years		2 years
[12]	IIIb 64	58	74	46	67	27	59*	73	81
					1.5 years
[13]	II 7, III 43			50	70
[14]	IIb 28, IIIb 92	48	72*
					3 years				3 years
[10], [11]	IIb 22, IIIa 1, IIIb 80, IVa 11	57	83*	41	61*			27	51*
					3 years				3 years
[15]	IIIb 40	50	80*	49	80*			48	58
					3 years				3 years
[16]	IIb 56, III 51, IV 3			∼80	∼70			73	73

RT = radiotherapy, HT = hyperthermia; * = significantly better than RT alone with p < 0.05.

Datta et al. [12] used 27 MHz capacitive heating with external electrodes and treated patients twice weekly, before RT, with an intra-vaginal temperature of 42.5°C for 15–20 min. Sharma et al. [13] used capacitive heating with an intra-vaginal electrode. Patients received three treatments per week, 42–43°C at the tumour surface for 30 mins, before RT. The publication of Hong-Wei et al. [14] is in Chinese, which makes most of the data inaccessible. They randomized patients to four treatment groups: RT alone, RT plus HT, RT plus chemotherapy (cisplatin, 5FU and vincristin) and RT plus HT plus chemotherapy. In Table I, the comparison between treatments with and without HT is shown. Harima et al. [15] used the 8 MHz Thermotron capacitive system for heating, using external electrodes. Patients received three treatments of 60 min at an average temperature of 40.6°C, after RT.

In the study by Vasanthan et al. [16], five institutes from various countries participated. Eight MHz capacitive heating systems were used to induce HT. For at least half of the patients, an intra-vaginal electrode was used. Five-to-seven HT treatments of 60 min duration were given before or after RT at an average temperature of 38.1–42°C.

Discussion

The DDHT clearly demonstrates that the addition of HT to conventional RT can result in large beneficial effects. These results were considered sufficiently convincing to start using HT as part of regular care for patients with advanced cervical cancer. This was not just because of the large and highly significant differences observed in the final analysis, but also because similar differences had been seen during the course of the trial and within the two separate studies which were combined in this trial. Furthermore, the Asian studies published earlier had shown similar beneficial effects from HT.

The acceptance of using HT in standard treatment for advanced cervix cancer was eased by economical considerations. There is a widely spread misconception that HT is an expensive treatment. It is true that application of the treatment is labour-intensive and, therefore, requires the availability of some respectable resources. Nevertheless, when the financial aspects are considered in terms of cost-effectiveness, as was done in this trial, HT appears highly cost-effective. The discounted cost-per-life-year-gained of Euro 3956, with the conservative assumption that a difference in eventual overall survival would be 11%, fits well within the range of accepted other medical interventions [17].

The Dutch study is the only trial in which a significant improvement in overall survival was demonstrated. The Asian studies show a significant benefit only for complete response and/or local tumour control. Two of the three studies reporting on overall survival, however, do show a trend of better overall survival by addition of hyperthermia to radiotherapy. The lack of benefit from HT in the study by Vasanthan et al. may be partly explained by an imbalance in tumour volume over the two treatment arms, with larger tumours in the RT + HT arm. Probably more important in their study is that the treatment application method has been inadequate [18]. They have used a capacitive heating technique, with which it is principally possible to achieve therapeutic heating in depth, provided that the limitations of this technique are taken into account. The one centre that treated half of the patients, and possibly other centres as well, report to have used an intra-vaginal electrode. This results in a high energy level and high temperatures in a small volume around the intra-vaginal electrode with much lower temperatures in the periphery of the tumour. A further limitation is related to excessive subcutaneous fat heating, which can be counteracted by selecting patients with a fat thickness of less than 2 cm and by pre-cooling the skin and subcutaneous fat. In the study by Vasanthan et al. patients were eligible when the subcutaneous fat thickness was up to 3 cm and pre-coooling is reported by only one centre. Therefore, subcutaneous fat heating may have been a limiting factor for sufficient energy deposition in the target volume. This assumption appears reasonable when the applied power level is considered. This, reported by one centre, was less than half the levels of those applied by Harima et al. [15], who did find beneficial effects of hyperthermia.

Conclusion

The main conclusion of the hyperthermia trials is that for patients with advanced cervical cancer, both pelvic tumour control and overall survival can be improved by the addition of hyperthermia to standard radiotherapy. Since these improvements were not accompanied by an enhancement of radiation-induced toxicity and the toxicity from hyperthermia was minor and of little clinical relevance, addition of hyperthermia clearly results in a therapeutic benefit. In the Netherlands, hyperthermia has become part of standard treatment of cervical cancer. In countries where this is not the case, hyperthermia should be given to those patients who cannot tolerate cisplatin treatment.